Predictors of Stroke Recurrence After Initial Diagnosis of Cervical Artery Dissection
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Objective
Patients with cervical artery dissection (CAD) are at increased ischemic stroke risk. We aimed to identify characteristics that are associated with increased risk of ischemic stroke following initial presentation of CAD and to evaluate the differential impact of anticoagulant versus antiplatelet therapy in these high-risk individuals.
Methods
This was a secondary analysis of the Antithrombotic Treatment for Stroke Prevention in Cervical Artery Dissection (STOP-CAD) study, a multicenter retrospective observational study. The primary outcome was subsequent ischemic stroke by day 180 after diagnosis. Patient characteristics were compared between those with vs. without subsequent ischemic stroke. Significant predictors were identified using stepwise Cox regression. Associations between subsequent ischemic stroke risk and antithrombotic therapy type in high-risk patients were explored using adjusted Cox regression.
Results
4,023 patients (mean age 47.4 years; 44.5% were women) were included. By day 180, 5.3% experienced a subsequent ischemic stroke. In adjusted Cox regression, factors associated with increased subsequent ischemic stroke risk were prior ischemic stroke (aHR 7.31, 95% CI 1.61-33.13, p=0.010), presentation within seven days from symptoms, (aHR 3.04, 95% CI 1.04-8.91, p=0.043), infarct on imaging (aHR 9.85, 95% CI 3.65-26.58, p<0.001), and occlusive dissection (aHR 2.34, 95% CI 1.03-5.34, p=0.043). Only patients with occlusive dissection had reduced subsequent ischemic stroke risk with anticoagulation versus antiplatelets (HR 0.37; 95% CI 0.15-0.89, p=0.03).
Interpretation
This study identified several predictors of subsequent ischemic stroke among patients with CAD but only patients with occlusive dissection demonstrated a benefit from anticoagulation. These findings require validation by meta-analyses of prior studies.
